The antennas used in the Synthesis Telescope have equatorially mounted, paraboloidal reflectors. There are two slightly different designs in use, the most significant difference being that two antennas (at the east and west extremes of the array) have diameter d=9.14 m and focal length f=3.81 m (f/d=0.42), while the remainder have d=8.53 m and f=3.66 m (f/d=0.43). The difference in antenna diameters leads to a mismatched weighting of u-v samples which is sufficiently large to affect imaging; Willis (1999) has devised a means of compensating for this effect.
Signals are collected by a dual-frequency feed (Veidt et al. 1985; Trikha et al. 1991) mounted at the prime focus. Circular polarization (CP) is received at both frequencies: continuum emission is expected to have a small fraction of linear polarization (from synchrotron radiation), and measurement of that component can be made most simply by using CP feeds.
At 1420 MHz a multi-mode feed, based on one of the designs of
Sheffer (1975),
provides a flat-topped illumination pattern, yielding
an aperture efficiency
,
with an edge illumination of the
reflector of -16 dB. A "quarter-wave plate'' in the waveguide
(consisting of three reactive posts;
Simmons 1952)
provides a 90
phase shift, and linear probes aligned at 45
to the plane of the
quarter-wave plate collect left-hand circular polarization (LHCP) and
right-hand circular polarization (RHCP) simultaneously.
The circularly polarized 408-MHz feed was added subsequently, and was
designed to avoid any degradation of 1420-MHz performance. The outputs of
four orthogonal monopoles, placed in the outer cavity of the 1420-MHz
feed, are combined in a TEM-mode hybrid. A switch selects either LHCP or
RHCP; normally RHCP is received (the use of a CP feed gives an accurate
measurement of Stokes I even in the presence of some linearly polarized
emission). At 408 MHz the feed has the radiation characteristics of a
circular waveguide of diameter
,
illuminating the aperture very
broadly. The aperture efficiency is
,
but edge illumination is
high (-8 dB) and beam efficiency is low (
)
because of the
high spillover.
Antenna noise temperature at 1420 MHz varies from a high of 26.0 K on one antenna to a low of 17.8 K on another, because of slight differences in construction. Table 2 gives details of the noise budgets of the best and worst antennas. These data were derived by Anderson et al. (1991) based on two-dimensional mapping of the radiation pattern of one of the antennas. The feed-support struts of several antennas have been modified (Landecker et al. 1991) to reduce the ground noise which the struts scatter into the aperture.
Cosmic microwave background | 2.7 | ||
Galactic emission | 1.0 | ||
Atmospheric emission (zenith) | 2.0 | ||
Main beam contribution: | 5.7 | ||
Worst | Best | ||
antenna | antenna | ||
Spillover | 8.0 | 8.0 | |
Diffraction at reflector rim | 0.6 | 0.6 | |
Mesh leakage | 5.9 | 1.5 | |
Strut scattering | 5.8 | 2.0 | |
Sidelobe contributions: | 20.3 | 12.1 | |
Total: | 26.0 | 17.8 |
Copyright The European Southern Observatory (ESO)